Titration of submonolayer Au growth on Si(111)

J. Kautz, M. W. Copel, M. S. Gordon, R. M. Tromp, and S. J. van der Molen

Phys. Rev. B 89, 035416 – Published 14 January 2014

Abstract

We study and analyze the growth of submonolayers of Au on Si(111) by a complementary set of surface techniques. Specifically, we focus on the $5 \times 2$ and the $α \sqrt{3} \times \sqrt{3}$ structures. We determine the gold coverage of these structures as a function of temperature by means of low energy electron diffraction (LEED) and low energy electron microscopy (LEEM). These results are independently calibrated by ex-situ ion scattering experiments. This allows us to present a phase diagram for this system. Remarkably, for all temperatures considered (820–1040 K), we find a coverage for the $5 \times 2$ phase that is significantly ( $\approx 10 %$ ) higher than the value of 0.6 monolayers which is assumed in the latest structural models. Therefore, a further refinement of the present picture of the quasi-one-dimensional $5 \times 2$ structure is required.

Received 7 October 2013

DOI:https://doi.org/10.1103/PhysRevB.89.035416

Authors & Affiliations

J. Kautz

Kamerlingh Onnes Laboratory, Leiden University, Niels Bohrweg 2, 2300 RA Leiden, The Netherlands

M. W. Copel and M. S. Gordon

IBM Research Division, Thomas J. Watson Research Center, Yorktown Heights, New York 10598, USA

R. M. Tromp

IBM Research Division, Thomas J. Watson Research Center, Yorktown Heights, New York 10598, USA and Kamerlingh Onnes Laboratory, Leiden University, Niels Bohrweg 2, 2300 RA Leiden, The Netherlands

S. J. van der Molen^*

Kamerlingh Onnes Laboratory, Leiden University, Niels Bohrweg 2, 2300 RA Leiden, The Netherlands

^*Corresponding author: molen@physics.leidenuniv.nl

Article Text (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 89, Iss. 3 — 15 January 2014

Reuse & Permissions

Access Options

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
Schematic of the aberration corrected low energy electron microscope. The red trace shows the electron path from the gun via the sample and the mirror towards the detector, where an image is formed. The electrostatic mirror corrects for the aberrations of the objective lens. We can switch between real space (LEEM) and reciprocal space (LEED) imaging.
Reuse & Permissions
Figure 2
(a) LEEM video frames from an evaporation experiment at 920 K in bright-field imaging mode. Upon gold evaporation onto a bare $7 \times 7$ Si(111) sample, islands with the $5 \times 2$ structure are formed (bright in the first two frames). After the $5 \times 2$ islands cover the entire sample at time t1, we change the electron energy from 14.5 to 8.2 eV to optimize contrast between $5 \times 2$ (dark) and $α \sqrt{3} \times \sqrt{3}$ islands (bright). (b) Surface coverages of the $5 \times 2$ and $α \sqrt{3} \times \sqrt{3}$ reconstructions versus time during gold evaporation. The fitted blue line shows the growth and decrease of the $5 \times 2$ coverage while the dashed green line gives the coverage of the $α \sqrt{3} \times \sqrt{3}$ reconstructed area. Evaporation starts at $t = 0$ . The first $5 \times 2$ ( $α \sqrt{3} \times \sqrt{3}$ ) islands become visible at $t_{0}^{'}$ ( $t_{1}^{'}$ ) and completely fill the surface at $t_{1}$ ( $t_{2}$ ). (c) LEED pattern of the $5 \times 2$ and $α \sqrt{3} \times \sqrt{3}$ reconstructions taken at time $t_{1}$ and $t_{2}$ , respectively. The dark shadows in both images are caused by secondary electrons. (d) MEIS results depicting the number of counts as a function of scattering energy for two samples with a complete $5 \times 2$ and $α \sqrt{3} \times \sqrt{3}$ coverage, respectively, grown at 920 K. By comparing the integrated area under the gold reflection peaks, we obtain the ratio between the two gold coverages. We find $Θ_{5 \times 2} / Θ_{α \sqrt{3} \times \sqrt{3}} = 0.65 \pm 0.01$ .
Reuse & Permissions
Figure 3
Experimental phase diagram for submonolayers of gold on Si(111). The cyan diamonds represent the MEIS results. The red squares are deduced from the times obtained in the evaporation experiments. The filled squares indicate the completion of the $5 \times 2$ and $α \sqrt{3} \times \sqrt{3}$ phase, corresponding to $t_{1}$ and $t_{2}$ , respectively. The open squares represent the first observation of domains of the $5 \times 2$ and $α \sqrt{3} \times \sqrt{3}$ phase, corresponding to $t_{0}^{'}$ and $t_{1}^{'}$ . For both the MEIS experiment and the growth experiment the value for completion of the $α \sqrt{3} \times \sqrt{3}$ structure was fixed to 1.00. The blue circles are the results from the temperature variation experiments in Fig. 4. The green and pink triangles depict the transition of the $α \sqrt{3} \times \sqrt{3}$ and $5 \times 2$ phase, respectively, into the $1 \times 1$ phase, for coverages between $Θ_{5 \times 2}^{'}$ and $Θ_{α \sqrt{3} \times \sqrt{3}}$ . For lower and higher coverages, the estimated melting temperatures have been depicted as smooth transitions in the phase diagram.[10] Our results show that the pure $5 \times 2$ phase has a finite width that increases with temperature. Above 920 K we only know the high coverage boundary of the $5 \times 2$ coverage. The hatched area depicts how we expect the $5 \times 2$ phase to continue.
Reuse & Permissions
Figure 4
Temperature dependence of the $5 \times 2$ coverage for three samples with different starting coverages for the $5 \times 2$ and $α \sqrt{3} \times \sqrt{3}$ reconstructions. Filled and open markers correspond to measurements after respectively raising and lowering the temperature. The lines are guides to the eye. As we increase the temperature the $5 \times 2$ coverage reduces. Since the total amount of gold on the sample remains constant, the number of gold atoms in the $5 \times 2$ and $α \sqrt{3} \times \sqrt{3}$ reconstruction, respectively, must be changing. Inset: FWHM of the $\sqrt{3} \times \sqrt{3}$ peak as a function of temperature. Around 980 K the $α \sqrt{3} \times \sqrt{3}$ structure melts into a disordered $1 \times 1$ phase.
Reuse & Permissions

Physical Review B

covering condensed matter and materials physics