This was an invited Perspectives from the editors of Science to comment about the paper of Markus Ternes and co-workers on atomic manipulation using AFM, trying to make it more accessible to non-specialized readers.
The work was a collaboration between two relevant groups in the international panorama of Scanning Probe Microscopy –the group at IBM Almaden (
These authors provide the closest measurements technically possible (at present) of the vertical and lateral threshold forces required for the manipulation of single atoms and molecules on a metallic surface. They used a dynamic force microscopy implementation working with very small oscillation amplitudes, so that they were able to correlate short-range chemical interaction forces with reliable measurements of the tunneling current flowing between tip and surface.
Surprisingly, they found that the lateral force to pull a Co atom on a Pt(111) surface seems to be independent of the vertical force –at least in the tip-surface distance range explored by the authors. This behavior contrasts with the common knowledge of the macroscopic world, in which the fiction force –the force to pull or push an object– is proportional to the applied load.
They also provide maps of the tip-surface interaction potential, fabricated from vertical forces obtained by scanning the surface at constant height at distances above and below the manipulation threshold –in the later case the atom is dragged by the tip visiting all the possible surface adsorption positions. These potential maps provide information about the adsorption stability of the atom (or molecule) on the surface. The authors report energy values for the saddle points connecting adjacent stable adsorption sites that are in surprisingly good agreement with STM diffusion experiments and first-principles calculations. This similarity in energy barriers is quite astonishing as the tip-surface interaction potential in principle differs from the surface potential energy felt by the atom. These magnitudes might be, however, alike upon an atomically rigid tip apex and a stronger interaction of the adsorbate with the tip. At the most instable adsorption sites, forces –and therefore potentials– are not well defined due to atomic jumps that give rise to an apparent dissipation of energy from the cantilever oscillation.
In this Science paper, we report a new atomic manipulation method that is based on the controlled and reproducible vertical interchange of atoms between the tip of an atomic force microscope (AFM) operated in dynamic mode and a semiconductor surface.
At variance with previous methods, these manipulations were produced by exploring the repulsive part of the short-range interaction between the closest tip-surface atoms.
Using this new manipulation method, we demonstrate that it is possible to assemble complex atomic patterns at a semiconductor surface by literally “writing with atoms” (depositing one atom at a time) in a very favorable time scale and at room temperate. See, for instance, the construction of the symbol associated with the silicon element (Si) in the animated GIF on the right by depositing silicon atoms in a perfect overlayer of tin atoms grown on a Si(111) substrate. We use the AFM as a pencil tool that can plot and erase by alternately depositing two atomic species at a heterogeneous surface.
This manipulation technique differs from methods previously reported using Scanning Tunneling Microscopy (STM), in which an atom on a metallic surface can be reversibly transferred between the tip and the surface by applying an appropriate bias voltage (see Don Eigler's paper in Nature). It also diverges from other methods of controlled atom manipulations recently achieved with AFM (see our Nature Materials paper and Markus Ternes and co-workers Science paper) that make use of the attractive part of the tip-surface interaction to laterally manipulate atoms without any active participation of the tip, which is only used to tune the interaction of the manipulated atom with the surface.
To gain deeper understanding on the mechanism of these manipulations, we have performed atomistic simulations working at the limit of current first-principles methods capabilities. We have found that the vertical interchange of atoms is controlled by the mechanical properties of a hybrid tip-surface dimer-like structure formed in the repulsive regime of the tip-surface interaction force. We have also estimated the energy barriers between the relevant atomic configurations that lead to these manipulations; barriers that are low enough to favor these manipulation processes at room temperature.
The results reported in this paper provide evidence that AFM can be used to incorporate individual dopants in semiconductor surfaces following predetermined and complex atomic arrangements. Although we focus on a tin/silicon alloy, we have reproduced these atomic manipulations in other semiconductor surfaces.
This manipulation technique may pave the way toward selective semiconductor doping, practical implementation of quantum computing, or atomic-based spintronics. The possibility of combining sophisticated vertical and lateral atom manipulations with the capability of AFM for single-atom chemical identification (see our Nature paper) may bring closer the advent of future atomic-level applications, even at room temperature.