Tải bản đầy đủ - 0 (trang)
Visualising (intra)molecular force-fields and submolecular structure

Visualising (intra)molecular force-fields and submolecular structure

Tải bản đầy đủ - 0trang

Since Gross et al.’s striking data were published, both the IBM group and

a number of other research teams have produced high resolution NC-AFM

images (using the qPlus variant of the technique, for the reasons discussed in

Section 2). An important review written by Leo Gross was published in

Nature Chemistry in March 2011.9 I will therefore largely forego a discussion of the results presented in that review and instead focus in this

section on developments since its publication.

Before moving on to cover recent examples of intramolecular resolution

using NC-AFM, however, it is essential that I highlight a novel approach to

high resolution STM imaging whose discovery pre-dates the publication of

the IBM pentacene work. This is the scanning tunnelling hydrogen microscopy (STHM) technique introduced by Temirov et al.53–56 Included in Fig.

10(b) is an image of PTCDA (3,4,9,10-perylenetetracarboxylic-dianhydride)

on a Au(111) surface, acquired using STHM and where it is clear that the

image resolution is comparable to that attained in the NC-AFM data.

This is largely because the two techniques share the same key elements – a

passivated tip (H-passivated in the case of Termirov et al.’s results) and

operation within the Pauli repulsion regime – although the contrast

mechanisms are of course rather different. For STHM, the passivated tip

translates variations in tip-sample force into a modulation of the junction

conductance. As can be seen from Fig. 10(b) the image resolution far

exceeds that observed in conventional STM because the latter is sensitive

only to variations in electron density in a relatively narrow energy window

close to the Fermi level. In STHM, as pointed out by Temirov et al.,55 it is

variations in the total electron density – and the information on chemical

structure embedded within it – which are probed.

On the basis of a series of combined Green’s function-local orbital density

functional theory calculations, Martı´ nez et al.57 have very recently put

forward a slightly different argument regarding the attainment of ultrahigh

resolution in STHM. They propose that hydrogen molecules are dissociated

at the tip and that the H atoms dramatically modify the density of states

(DOS) at the Fermi level. It is this modification of DOS at EF which they

claim gives rise to the enhanced resolution. It should be noted, however,

that Temirov and colleagues53 previously listed a number of experimental

observations to support their claim that it is molecular, rather than atomic,

hydrogen (or deuterium) that underpins the STHM imaging mechanism.

If it is indeed H atoms, rather than H2 molecules, at the tip which are

responsible for the intramolecular contrast in STHM then there is the

exciting potential to combine STHM and NC-AFM measurements of

molecules adsorbed on the H:Si(100) (or H:Ge(100)) surface by exploiting

the passivated state of the tip which results simply from scanning the surface, as described in Section 3.3 above. This in turn could provide important

new insights into the relationship between molecular conformation and

function for molecular logic gates on semiconductor substrates.

The precise atomic and electronic structure of the tip apex of course

underpins all SPM and over the last couple of years there has been rapidly

growing interest in characterising and controlling the probe state to a much

greater extent than ever before (which I return to in Section 6). Before

leaving STM to focus on recent NC-AFM work, I would like to briefly

130 | Nanoscience, 2013, 1, 116–144



Fig. 11 STM images of a naphthalocyanine molecule taken with (a) a Cu (i.e. ‘s wave’) tip, and

a CO (i.e. ‘p wave’) tip in (b) constant current, and (c) constant height mode. Note the much

higher resolution for the CO-terminated tip due to its sensitivity to the gradient of the wavefunction. Taken from Ref. 19.



discuss an elegant set of results from Gross et al.19 involving an analysis of

the role that the tip wavefunction plays in high resolution imaging of submolecular structure (and, more broadly, in scanning tunnelling microscopy

in general). This work is perhaps best explained in the context of Fig. 11

which shows the difference between STM images of a napthalocyanine

molecule (adsorbed on a thin NaCl film so as to decouple it from the

underlying metallic substrate) taken using a Cu tip and a CO tip. Only the

latter involves tunnelling through px and py tip orbitals and, as pointed out

by Chen58 in the early years of the STM field, these p states provide access to

the spatial derivative of the sample wavefunction and, therefore, yield much

greater resolution. The Cu tip, by contrast and by virtue of its s orbital

character, produces images which are, in essence, a map of the local density

of states close to the Fermi level (or, more correctly, within an energy

window defined by the tip-sample bias) – the ‘traditional’ STM approach.

While STM is sensitive to the details of the frontier orbitals (i.e. the

energetically low-lying molecular wavefunctions), NC-AFM is capable of

not only resolving atoms but, under the appropriate choice of imaging

conditions, charge density variations due to interatomic/intramolecular

bonds. Quite how atomic resolution is achieved, however, has been a matter

of some debate. Perhaps somewhat surprisingly, Gross et al. made the

following claim in their pioneering paper on the attainment of submolecular

resolution of pentacene: ‘‘We conclude that atomic resolution in NC-AFM

imaging on molecules can only be achieved by entering the regime of repulsive

forces because vdW and electrostatic forces only contribute a diffuse attractive

background with no atomic-scale contrast.’’10 This strategy of imaging within

the Pauli exclusion regime of the tip-sample potential in order to provide the

highest levels of intramolecular contrast is increasingly being exploited by

NC-AFM researchers. Figure 12 is a striking recent example taken from the

Nanoscience, 2013, 1, 116–144 | 131



Fig. 12 High resolution images of dibenzthianthrene molecules adsorbed on an ultrathin layer

of NaCl acquired using NC-AFM. Note that the raw data have been processed with a

Laplacian filter to enhance the curvature of the image and thus highlight the intramolecular

structure. (Ball-and-stick models of the molecules have also been overlaid on the image). The

inset is an atomic resolution image of the NaCl substrate. From Ref. 59.



work of Jascha Repp’s group at the University of Regensburg which shows

the intramolecular ‘architecture’ of two different adsorption states of the

dibenzo[a,h]thianthrene molecule which exists in two stable isomeric

forms.59

A rather different conclusion regarding the attainment of atomically

resolved images of molecular structure was reached by Ondracek et al.,60

however. They determined, on the basis of first principles calculations, that

atomic resolution was indeed possible in the attractive regime of the tipsample potential and that the critical parameter was the tip reactivity. Their

results echo those of Hobbs and Kantorovich from a number of years ago

whose DFT calculations strongly indicated that atomic resolution in NCAFM was possible for a C60 molecule scanned by a silicon tip.61

In order to experimentally elucidate the role of the scanning probe in

atomic resolution STM or NC-AFM imaging, one of course needs to know

the precise state of the tip. This is fraught with difficulty because tip preparation in many cases is one of the scanning probe microscopist’s ‘guilty

secrets’ – a relatively small proportion of tips work (i.e. give atomic resolution) on the first scan (and, even if they do, that capability can be lost and

restored very many times in the lifetime of a probe). Instead, probe microscopists invest a significant amount of time and effort into coercing the tip

into the ‘correct’ state largely on a trial-and-error basis (see Section 6 below).

Very important advances in controlling the precise structure of the tip apex

have, however, been made by Guillaume Schull, Richard Berndt and coworkers in a series of elegant experiments focussed, in the main, on measuring

the conductance of molecular junctions.62–65 Figure 13 shows a number of

examples of the degree of control and the extent of probe characterisation

attained by Schull et al. where atoms adsorbed on a Ag(111) surface are used

to ‘inverse image’ the tip structure, following controlled transfer of a C60

molecule to the apex. The Ag atoms, by virtue of their smaller ‘radius of

curvature’ compared to the fullerene at the end of the STM tip, image the

probe apex. This turns the conventional view of STM imaging on its head –

the surface is imaging the tip, rather than vice versa. But this is also entirely

consistent with the fundamental imaging principle at the heart of scanning

probe microscopy – a probe with a small effective radius of curvature images

132 | Nanoscience, 2013, 1, 116–144



Fig. 13 Controlling the orientation of a C60 molecule adsorbed on an STM tip. (a), (b) Differential conductance (dI/dV) spectra for a C60-terminated tip above a C60 molecule and the

metal substrate respectively; (c) – (e) STM images, and (f) – (h) dI/dV maps, for a C60 terminated tip above metal adatoms and clusters; (i) – (l) reorientation of the tip-bound C60 via the

application of high current. From Ref. 62.



a larger structure. Generally, the tip has the smaller radius of curvature but in

this case adsorbed atoms at the surface are exploited to do the imaging.

Over the last decade Giessibl et al.66,67 extended this approach to its

logical limit by exploiting dangling bond orbitals at the Si(111)-(7 Â 7)

surface as ‘mini-tips’ which could be used to image the orbital structure – or,

more correctly, map the tip-sample interactions arising from the orbital

structure (orbitals are not, of course, an experimental observable) – of

a scanning probe. Building on both Giessibl et al.’s and Schull and

co-workers’ work on tip apex control and characterisation, we recently

obtained atomic resolution images of a C60 molecule at the end of an STM/

NC-AFM tip (Fig. 14(a)).68 The five-fold symmetry of a pentagonal face of

the fullerene molecule is clearly resolved, as is the effect of the molecular tilt

at the end of the tip – one atom is clearly interacting more strongly with the

surface than the others. In addition, there is an asymmetry in the contrast of

the five-lobed features in the two sides of the (7 Â 7) unit cell, arising from

the well-known difference in the electron density and, thus, chemical reactivity of the adatoms in the unfaulted vs. faulted halves of the cell. Also

included (Fig. 14(b)) is a force-distance spectrum which highlights that

submolecular (atomic) resolution in this case is due to a weakly attractive,

rather than a repulsive, interaction. The reactivity stems from the partially

filled nature of the adatom dangling bond which interacts strongly with the

fullerene cage, forming an (iono)covalent bond.

A wide variety of molecular tilts/orientations have been observed

experimentally and can be characterised and catalogued using both atomic

resolution NC-AFM and (dynamic) STM. To interpret images acquired

with the latter technique we use the Huckel orbital approach developed by

Dunn and co-workers69 as a low computational cost protocol for

Nanoscience, 2013, 1, 116–144 | 133



Fig. 14 Atomic resolution imaging of a tip-adsorbed C60. (A) The dangling bond orbitals of the

Si(111)-(7 Â 7) surface are used as an array of ‘mini-tips’ to image the structure – and, thus,

ascertain the precise orientation – of the fullerene molecule. In this case the pentagonal face of the

molecule is pointing towards the surface and there is a clear molecular tilt so that one C atom

appears much brighter in the image than the other four. Note also the difference in contrast between

the two sides of the unit cell. (B) Force vs. distance curve which shows that the maximum attractive

force due to Si-C bond formation is B 1.6 nN. (C) Imaging in a pseudo-constant-height mode

above an adsorbed C60 molecule (with a C60-terminated tip) highlights intramolecular contrast due

to the carbon-carbon bonds. A high pass filter (D) accentuates the contrast. From Ref. 68.



ascertaining molecular orientation. In more recent work we have focussed

on extending this approach to molecule-on-molecule imaging with submolecular resolution. (Berndt et al. are also making important in-roads in

this area.) The passivated nature of the C60-terminated tip also enables

imaging in the Pauli exclusion regime of the probe-sample interaction

potential – in the manner introduced by Gross et al.10 – and an example of

the contrast obtained via this approach is shown in Fig. 14(c). Note the

narrow width of the submolecular features.

Terminating the apex of a scanning probe with a single molecule enables

intermolecular potentials to be measured on a molecule-by-molecule basis.

Two examples from the recent literature are shown in Fig. 15. Early in 2011,

Sun et al.70 reported the measurement of the CO-CO interaction potential

using a CO-functionalised tip above an adsorbed CO molecule. Figure 15(a)

is a comparison of their experimental interaction potential against that

derived from DFT calculations. The agreement is remarkably good; deviations from the calculated curve in the repulsive regime of the potential arise

from relaxations of the CO molecule at the tip but, as the authors themselves

point out, the close correspondence of the experimental data with the calculation for two free (i.e. unadsorbed) CO molecules is surprising.

Similarly good agreement between experiment and theory – in this case,

an analytically derived, rather than DFT, prediction – is observed for the

C60-C60 potential shown in Fig. 15(b). The Girifalco potential71 accurately

models our experimental data both in terms of the depth and the width of

the intermolecular interaction curve. Deviations at intermolecular

134 | Nanoscience, 2013, 1, 116–144



Fig. 15 Intermolecular potentials measured using NC-AFM. (a) CO-CO. Note the particularly good agreement between density functional theory and experiment for the attractive

regime of the potential. (The inset shows a zoom of the region around the minimum of the

potential). ‘Snapshots’ of the state of the CO-CO junction at various tip-sample separations are

shown underneath the graph. From Ref. 70. (b) C60-C60. In this case an accurate analytical

potential exists and there is again very good agreement with experiment around the minimum

of the potential. Note that no fitting has been carried out – the experimental curve has simply

been shifted along the x axis (i.e. in Z) to match the position of the minimum in the Girifalco

potential. From Ref. 68.



separations smaller than the equilibrium separation arise from molecular

relaxations at both the tip and the sample surface and are to be expected –

the Girifalco potential takes no account of this type of relaxation. The

theoretical potential systematically overestimates the C60-C60 interaction for

intermolecular separations which are more than B1 A˚ larger than the

equilibrium value. Again, this is expected because the experiment deals with

molecules which, far from being the ‘free space’ entities considered by

theory, are adsorbed on a tip and a Si(111)-(7 Â 7) surface. This will modify

both the dipole moment of the molecule and, importantly, its polarisability.

The experimental C60-C60 potential of Fig. 15(b) therefore provides an

interesting, but rather challenging (due to the system size), test case for DFT

codes which incorporate dispersion interactions.

It is also possible to carry out force-distance spectroscopy across a grid of

(x,y) points above a molecule, enabling a 3D map of the interaction

potential to be generated. Particular care has to be taken, however, to

ensure that neither the tip-adsorbed molecule nor the ‘target’ molecule

adsorbed at the surface are disturbed. This is especially true if high resolution spectra within the repulsive regime of the tip-sample potential are

required. Gross et al.72 put forward an experimental protocol based around

a variable distance of closest approach of the tip which facilitates the

acquisition of high resolution 3D maps of the interaction force/potential

while suppressing measurement instabilities.

The spectra shown in Fig. 15 were acquired with a functionalised/passivated probe, where a molecule had been deliberately transferred from the

surface to the apex of the tip. However, submolecular resolution has also

been reported for tips which the authors claim do not have a terminating

molecule.73,74 Two examples are shown in Fig. 16. The first is of particular

relevance to the C60 results discussed above and shows submolecular resolution of C60 adsorbed at a surface rather than on a tip. The authors have

Nanoscience, 2013, 1, 116–144 | 135



Fig. 16 Submolecular resolution achieved with tips which were not deliberately functionalised

with a molecule. (a) Intramolecular features observed for C60 molecules adsorbed in a thin film

on a Cu(111) surface; (b) Force-distance curve above a C60 molecule. The authors of Ref.

interpreted their results in terms of a Cu-C60 interaction; (c) Submolecular contrast of a decastarphene molecule. (a) and (b) are taken from Ref. 73, (c) and (d) from Ref. 74.



also measured the interaction potential between the tip and a C60 molecule

(Fig. 16(b)). There are three rather interesting, and somewhat surprising,

aspects of these results. First, submolecular resolution is achieved for

molecules not in the first adsorbed layer (i.e. directly interacting with the Cu

substrate) but on top of fullerene islands, where the intermolecular interaction is purely of van der Waals character and is rather weak. Second,

submolecular resolution showing carbon-carbon bond-derived features

within the fullerene cage is apparently achieved with a metal-coated

(Cu-terminated) tip. And, third, the tip-sample interaction force at the

minimum of the potential isB50 pN. This latter result is particularly intriguing given that the van der Waals interaction between two C60 molecules is

approximately six times larger and that, at least for C60 adsorption on

Cu(111) surfaces, the fullerene-metal interaction involves some degree of

charge transfer. In addition to submolecular resolution imaging of C60 using

a Cu-terminated tip, intramolecular contrast in the absence of a functionalized probe has also been reported for a decastarphene molecule adsorbed

on both the Cu(111) surface and on a 2ML film of NaCl on Cu(111)75

(Fig. 16(c)).

In the weeks before submission of this review, Wagner et al.76 reported a

protocol for extracting the adsorption energy of organic molecules on metal

surfaces from NC-AFM df(z) spectra. This involved a sophisticated and

comprehensive exploration of the multi-dimensional parameter space

136 | Nanoscience, 2013, 1, 116–144



involved in fitting the experimental data with a combination of empirical

potentials to account for van der Waals interactions, Pauli repulsion, and

chemical bonding. The authors argue that their approach enables the

relative contributions of each of these types of molecule-substrate interaction to be extracted from the experimental data.

As noted above, NC-AFM has the capability to produce 3D maps of the

tip-sample interaction. Welker and Giessibl77 exploited this to map the

angular dependence of both the tunnel current and the force between a CO

molecule adsorbed on Cu(111) and a metal (tungsten) tip. They found that

the force maps – and, to a much lesser extent, the (suitably normalised)

tunnel current – demonstrated clear angular dependencies which arose from

the geometric structure, i.e. the crystal symmetry, of the tip. This dependence allowed the authors to construct a semi-empirical potential based

around a combination of a Morse law and a much shorter range component

which was peaked in specific crystal directions (specifically, along the

o100W vectors). The o100W dependence and, indeed, the exponential

decay of the short-range component were rationalised in terms of the

interaction of the W electrons with the dipole of the adsorbed CO molecule.

NC-AFM is not just sensitive to chemical bonding and van der Waals

forces, however. The technique, in the form of the Kelvin probe force

microscopy (KPFM) variant, can also be a sensitive probe of electrostatic

potentials. When this electrostatic sensitivity is coupled with submolecular

resolution, remarkably powerful insights into charge distributions within

molecules become possible. This was elegantly demonstrated by Fabian

Mohn and co-workers by combining STM, NC-AFM, and KPFM measurements of a single naphthalocyanine molecule adsorbed on a thin

(2 monolayer) NaCl film on Cu(111) (Fig. 17).78 The charge distribution

within the naphthalocyanine is clearly resolved in the KPFM image for two

different tautomerization states of the molecule.

The charge distribution within the naphthalocyanine is clearly resolved in

the KPFM image for two different tautomerization states of the molecule.

As Mohn et al. highlighted in the conclusion of their paper, and echoing a

point made at the start of this review, while STM is sensitive to the density

of states of the frontier orbitals within a relatively narrow energy window

about the Fermi level, and AFM is a probe of total electron density, KPFM

provides maps of the electric field arising from the charge distribution

within a molecule (or, more generally, at a surface). The combination of

STM, AFM, and KPFM is thus an exceptionally powerful ‘toolbox’ for the

analysis and manipulation of matter at the atomic and (sub)molecular

levels.

5



‘Dialling in’ dirac fermions and addressing atomic spins



A remarkable ‘tour de force’ demonstration of the power of SPM-actuated

molecular manipulation was reported by earlier this year.79 As an example

of the impresive levels of control which are now possible using STM as a

molecular positioning tool, the ground-breaking and innovative work of the

Manoharan Group (Stanford) shown in Fig. 18 has ‘set the bar’ very high

for future research in the field. By laterally displacing CO molecules, one at

Nanoscience, 2013, 1, 116–144 | 137



Fig. 17 (a), (b) Kelvin probe force microscopy images of a naphthalocyanine molecule in two

different isomerisation states; (c) Difference map obtained by subtracting image (b) from image

(a); (d) result of a density functional theory calculation of the asymmetry in the z component of

the electric field above a free naphthalocyanine molecule. Taken from Ref. 78.



Fig. 18 Fabricating artificial graphene via STM manipulation of CO molecules so as to define

the appropriate potential landscape for electrons in the Cu(111) substrate. The image on the

right shows a distortion of the CO positions in that lattice which mimics the effect of applying a

60 T magnetic field. Adapted from Ref. 79.



a time, the researchers built up a potential energy landscape for electrons in

the Cu(111) substrate which simulated that of the graphene lattice, artificially producing the Dirac fermions which are a signature of that material.

Not content with generating ‘molecular graphene’ in this way, Gomes et al.

138 | Nanoscience, 2013, 1, 116–144



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

Visualising (intra)molecular force-fields and submolecular structure

Tải bản đầy đủ ngay(0 tr)

×