|
An AFM height image is usually not a true scan of surface topography
but a map of measurement results influenced by a number of different
artifacts. Among them, there are unavoidable structural artifacts,
like
· Convolution with the finite size and shape of the probe
tip,
· Surface deformation caused by tip-sample forces,
and instrumental artifacts that characterize the quality of the AFM
system like noise arising from cantilever backside roughness and feedback
loop limitations. Other physical properties of the surface may also
affect the tip-sample interaction process and bring about additional
uncontrollable changes to the collected topography data.
The first step to understanding the influence of the different
artifacts consists in the separation of errors coming from different
sources. This essential issue might be investigated by means of
the Scanning Probe Microscopy simulation program by solving the
equations of motion of the probe assuming the tip-sample force interactions.
The current version of the simulator is an idealized instrument
with no noise and ideal feedback (instantaneous and with no inertia)
that helps to reveal the pure look of the structural features.
Using the simulator, it is possible to
· Understand the evolution of the images due to changes
of the parameters of the experiment, e.g. tip's radius or applied
force,
· Determine how the height image differs from the real
topography depending on the parameters.
An iteration procedure based on the simulation results can be used
to restore the unknown real topography from the scan data, and a
systematic analysis of this kind of experimental work might be helpful
on the way to a confident straightforward measurement technique.
To get more information about the simulator contact our support
team.
|
 |
Case study: Alkane layers
Normal alkanes CnH2n+2 on graphite, which
are in solid crystalline state at room temperature starting with
C18H38 and finishing C390H782
can be examined using both STM and AFM. In studies of alkanes, STM
shows incredible resolution allowing visualization of individual
chains. In AFM, one can see only lamellae, whose edges consisting
of mobile -CH3 groups are seen as dark lines. The challenge
is to get the chain resolution in AFM similar to the one observed
in STM.
The model of alkanes based on embedded spheres was used for AFM
simulations.
|
 |
| Fig.1. Simulation result of C390H782
alkane layer with Rtip=5nm, Asp=19 nm |
|
|
 |
| Fig.2. Simulation of the same alkane layer
with Rtip=10nm, Asp=19 nm |
|
|
|
The results show a need of high Z-sensitivity (better than 0.5A)
and a sharp probe (tip radius of 5 nm or less) to get resolution
similar to STM. It is also necessary to use light tapping conditions,
when the set-point amplitude, Asp, is close to the amplitude
of a non-interacting probe, A0, measured in the immediate
vicinity of the sample surface.
|
|
|
Case study: Polydiacetylene crystal
PDA crystals have been examined with AFM since 1991 when the atomic-scale
images of this compound were obtained in contact mode [1]. Similar
ambient-condition images of PDA in AM
and FM modes were obtained more than 10 years later [2, 3]. These
results and the increased interest in molecular-scale imaging in
dynamic modes stimulated the theoretical modeling towards the better
understanding of the nature of such images [4]. The main results
of the computer simulations of the AFM images of the molecular arrangement
on the bc crystallographic plane of PDA are presented in Figs. 3
together with the crystallographic structure of this plane and experimental
AFM patterns recorded in AM mode. The atomic arrangement on the
crystallographic bc plane of polydiacetylene (PDA) crystal [1] was
modeled using the X-ray data and choosing only topmost F- and H-
atoms of these macromolecules. The variations of the simulated patterns
in Figs. 3a-c, which are caused by an increase of the tip radii
and the tip-sample force, are similar to those found in the experimental
images in Figs. 3d-f.
|
|
|
 |
| Fig.3. Simulated AFM images of PDA crystal
obtained at different forces and with tips of different radius. The
pattern in (a) was obtained in the "zero-force" imaging
with the 5-nm tip; the pattern (b) - in the imaging with the 5-nm
tip at Asp = 19nm, and the pattern (c) - in the simulation
with the 15-nm tip at Asp=18 nm. Cross-sections along the
horizontal direction are shown underneath the images. Bottom row:
corresponding experimental AFM images of the bc surface of the crystal
obtained in AM mode. |
|
|
|
Of special interest is the AFM imaging of single molecular defects.
The crystallographic arrangement with one molecular defect (vacancy)
in the center is precisely reproduced in the image simulated for
the atomically sharp tip in cases of the "zero-force"
imaging and operation at elevated force, Fig. 4a. The presence of
this defect is also noticed in the image patterns obtained with
tips having a radius from 1 nm to 10 nm, and its signature even
broadens when larger tips are applied, Fig. 4c-d. The image patterns
in Figs. 4c-d simulated with the larger probes show the details
do not directly relate to the surface structure.
|
|
|
 |
| Fig.4. AFM simulation patterns of the bc
plane of PDA crystal with a missing molecule in the center. The images
were obtained at elevated tip-sample force (Asp=10 nm,
A0 = 20 nm) for tips with radii of 150 pm (a), 1nm (b),
5 nm (c) and 10 nm (d). Scan size is 80nm x 80nm. |
|
|
|
In the analysis of simulated images one should be aware not only
of a particular pattern in the image, but also of changes in lateral
dimensions and vertical corrugations of the structures. It is important
to understand if these changes are accessible with AFM microscopes
that have signal to noise limitations. The surface corrugations
of the crystallographic lattice and molecular defects measured by
AFM are larger than 2 A in the ideal case of the atomically-sharp
probe, decreasing with the increase of the tip radius. It becomes
around 0.5 A for the 1 nm tip and around 0.2 A for the 10 nm tip.
Therefore, for observations of these structures in practice one
needs to use a microscope with an extremely low noise level [5].
|
|
|
|
References:
1. Magonov S., Bar G., Cantow H.-J., Bauer H.-D.,
Muller I, Schwoerer M. Polym. Bull. 1991, 26, 223.
2. Klinov, D.; Magonov, S. Appl Phys Lett 2004, 84, 2697
3. Fukuma T, Kimura M., Kobayashi K, Matsushige K, Yamada H. Rev
Sci Inst 2005, 76, 1.
4. Belikov S, Magonov S. Jap J Appl Phys 2006, 45, 2158.
5. Magonov S. American Laboratory 2008, February 28, 33.
|
|
|
