We are using scanning tunneling microscopy (STM) to explore the electrical, structural and magnetic properties of a variety of molecular species.  We have developed an ultra-low current (down to below 100 fA) UHV STM for studies of the electronic spin states of single molecules.


                                               Au (111) surfaces  



                                            Graphite surfaces & superlattices



                                             Molecules on surfaces

    We are particularly interested in HOPG (Highly oriented pyrolytic graphite), and in the many intriciacies of performing STM on this material, which is one of the most commonly used substrates for STM.  Due to the weak coupling between adjacent layers of graphene, graphite has extremely low sliding friction, hence its use in lead pencils.  This propensity for sliding also means that the local stacking of sheets can be easily disrupted, and this can be directly probed using STM.  Through these sort of experiments, we can learn alot about the intriguing properties of this material.

    What happens when there is a mis-orientation of graphene layers?  We get a superlattice as shown below.  Here, we have shown two graphene sheets with atomic lattice constant, d, mis-oriented by an angle Q, which leads to the formation of an interference pattern (known as a Moiré pattern - well-known from optics) with the same symmetry, but a superperiod, D.  The relationship between the various quantities is given by
         


                 overlap -> 


            In an STM image this looks like:


      
 
    In this image, we can simulatneously observe the superlattice as well as the atomic lattice.  Note that the atomic lattice has triangular symmetry, whereas the arrangement of the carbon atoms is in fact hexagonal.  This has been a subject of great debate for over 25 years now, and is due to the AB (Bernal) stacking of the individual graphene sheets.  Essentially, the STM observes every second carbon atom (known as b-atoms), i.e. the ones which are above a vacancy in the layer immediately below.

    We have recently demosntrated that the STM tip can be used to modify the spacing between the layers locally due to the attractive forces between the tip and sample, and we can essentially decouple the top layer, so that acts like graphene.  This is observed in the image below where we have modified the tip-sample distance during a scan to decouple the top layer, whereby we see the triangular lattice in the upper part of the image, and the true hexagonal lattice in the lower part, whilst the top layer is decoupled.               

HOPG warps under the STM tip during scanning.  On the right image, for the bottom half, we have used the STM tip to decouple (lift) the top layer of graphene, and in the top half, we have reverted to the normal coupling.     See our paper in ACS Nano on the decoupling of graphene layers in HOPG using the STM tip     This can be done in a controlled way as shown below.  Here we have a region on a piece of graphite (HOPG) where there usual stacking of the graphene sheets has been disrupted, resulting in a slight rotational misorientation between the top few layers at the Basal plane.  This manifests itself as a superlattice , which is apparent on the right hand half of the images.  In the superlattice region, the average interlayer spacing is slightly larger than usual, resulting in a reduced electronic coupling.  In the image on the left, the typical atomic resolution image of HOPG is observed (displaying triangular symmetry, thus showing every second C-atom) everywhere.  In the image on the right, where we have brought the STM tip closer to the surface, we still observe the triangular lattice on the left half, but now obtain the true atomic lattice (honeycomb, hexagonal symmetry) on the right half.  Due to the reduced interlayer coupling on the right, the STM tip is able to lift the top layer there by a few tenths of a nanometre, enough to decouple it almost completely, so it appears like graphene.              No decoupling: triangular lattice                              partial decoupling: trianglar lattice on left, honeycomb on right Zoom-in on right-hand half of both images (image size ~ 0.8 nm x 0.8 nm):                     Triangular lattice          Hexagonal lattice  We can also simulate these images using a simple model which we developed (see publications list):             Triangular lattice                      Hexagonal lattice     We have also gone a step further and have carried out a study on the dependence of atomic mismatch between graphene sheets in graphite and the misorientation of the resulting superlattice domains. We have shown, through measurements of the atomic orientation of the top sheet and the angular misorientation between superlattice domains, that it is possible to compute the actual degree of atomic mismatch on the underlying graphene sheet in the case of a superlattice. We show that the odd-even transition is evident for superlattices with relatively small periodicity in the range 1–2 nm and less apparent for those with larger periodicity in the range 5–8 nm, and present a signature of the transition. We also demonstrated that the degree of interlayer coupling between graphene sheets depends on the extent of rotational mismatch in relation to interlayer spacing as has previously been predicted.                                                                                                   Defect-high region of HOPG showing                     FFT from superlattice (measured)         FFT from superlattice (calculated) two superlattices with different period                    showing satellite peaks associated          showing identical features stemming from two graphene sheets                        with superlattice & odd-even transition with different rotational mismatch. Different period also evident in FFT (inset)           See our paper in Phys. Rev. B on rotational disorder in graphite